TRANSACTIONS OF SECTION A. 631 



No 3 X by 25 was taken, and the focus = 12". The angle made with the axis 

 by the ray in Calcite varied from 2° 47' 26" to 1° 14' 10" for W. L. 2748 and 

 from 2°;36'26" to 1°9' 17" for W. L. 5607, and the spherical aberration of the 

 combination for W. L. 2748 was --051337, and for W. L. 5607 was --069809. 

 No. 5 would probably have covered better without introducing too much 

 double refraction. 



7. The Production of an Artificial Light of the same Character as Day- 

 light. By Arthur Dufton, M.A,, B.,Sc., and Walter M. Gardner, 

 Bradford Technical College. 



It is a matter of common experience that many colours alter in appearance 

 when seen by artificial light. 



The extent to which colours may vary under different illumination is perhaps not 

 commonly known, but is well illustrated by the range of dyed cloths exhibited. 

 Amongst other patterns, one which is green by daylight becomes red-brown by 

 gaslight ; a violet changes to purple ; a grey to heliotrope ; a shade of tan to a 

 brick red. Particularly striking is a pattern woven from specially dyed yarns, 

 which appears a uniform green colour by daylight, but which is figured by 

 gaslight. Seen by the light of the electric arc, the patterns show similar but less 

 marked changes. 



_ It may be of interest to indicate briefly how such peculiar changes of colour 

 arise. The colour of a body depends in the first place on the nature of the 

 incident light. In monochromatic red light a red appears much the same as in 

 daylight, but a yellow changes to red, a green is almost black, while blues and 

 violets become red. 



Gaslight shows a continuous spectrum from red to violet, but compared with 

 daylight is of a strong orange colour due to an excess of rays in the red, orange, 

 and j'ellow It does not, however, necessarily result that all colours appear 

 redder by gaslight. It is, indeed, well known that the majority of colours change 

 little by gaslight. This is due to the adaptability of the eye ; if the light becomes 

 redder, the eye becomes less sensitive to red ; if the light is deficient in gi-een, the 

 eye_ becomes more sensitive to green. Persons working by gaslight soon cease to 

 notice its intense orange colour. It results that a grey produced by mixture of 

 black and white appears grey under any illumination, and simple colours, such as 

 reds, oranges, and some greens giving light confined practically to one part of the 

 spectrum, undergo little change. 



Generally, however, the colour of a body is due to a mixture of light from 

 different parts of the spectrum. All violet colours are transparent, not only for 

 violet, but also for blue and red light ; all blues transmit not only blue, violet, and 

 green light, but also more or less red. Consequently, whenever a blue or violet is 

 used in the production of what is called by artists a 'tertiary' colour, the general 

 result is a colour having bright bands in diflerent parts of the spectrum. A 

 mixture of red, blue, and yellow to produce a neutral grey will show bright 

 bands in the red and green— complementary colours, resulting in a proportion of 

 white light. According to the exact position and intensity of these bands the 

 grey will become redder or greener or may even remain unchanged by gaslight. 



Generally colours become redder under artificial light. This is due not merely 

 to the redder character of artificial lights as compared with daylight, but to the 

 peculiar transparency of colouring matters for red light. Among reds and yellow, 

 we have many theoretically perfect colouring matters — a perfect yellow being one 

 having sharp absorption in the violet and blue, and perfect transparency for green, 

 yellow, orange, and red rays. A perfect blue would be transparent for violet, blue, 

 and green, and opaque for the rest of the spectrum. 



Apparently such a blue can only be obtained by means of cupric salts. All 

 other blue dyes and pigments we have examined agree in being more or less 

 transparent for red light. Even greens transmit some red. This peculiar trans- 

 parency of colours for red light is of primary importance in colour-matching. All 



